Chemistry of tetravalent plutonium and zirconium : hydrolysis, solubility, colloid formation and redox reactions

Abstract

The chemical properties of plutonium and zirconium are important in order to assess nuclear waste disposals with respect to isolation and immobilization of radionuclides. In this study, the hydrolysis, solubility and colloid formation of tetravalent plutonium and zirconium are investigated in 0.5 M HCl/NaCl solution using several complementary methods and the redox behavior of plutonium is investigated in acidic conditions as well. The solubilities of Pu(IV) and Zr(IV) are determined from the onset of colloid formation as a function of pH and metal concentration using LIBD (laser-induced breakdown detection). The investigation of the solubility of Zr(IV) is carried out at different concentrations (log [Zr] = -3 ~ -7.6) and in a wide pH range (pH = 3 - 9) yielding log K°sp(Zr(IV)) = -53.1 ± 0.5 based on the assumption that only mononuclear hy-drolysis species exist in solution. Comparing the present results with literature data, the solubilities of Zr can be split in two groups, a crystalline phase with lower solubility and an amorphous phase (Zr(OH)4(am)) with higher solubility. The data obtained in the present work set an upper limit for the solubility of freshly formed Zr(OH)4(am). To understand this difference of solubilities, the geometrical structure of the dominant solution species is investigated as a function of pH using XAFS (X-ray absorption fine structure). The samples at pH >2, still below the solubility limit determined by LIBD, contain the polynuclear Zr(IV) species probably due to the high concentration ([Zr] = 1 mM) and their structure do not resemble any reported simple ZrO2 structure. The Zr(IV) colloid species in oversaturated solution under this experimental condition resembles amorphous Zr(IV) hydroxide rather than crystalline ZrO2. The solubility of Pu(IV) is investigated in acidic solution below pH 2. Considering only mononuclear hydrolysis species, log K°sp(Pu(IV)) = -58.3 ± 0.4 is obtained. Since Pu(IV) is not redox stable even in acidic condition, the concentration of each oxidation state of Pu must be determined prior to each experiment. The solubility data are determined directly after preparation and then the redox reactions between four different plutonium oxidation states are observed at different pH and Pu concentrations as a function of time. The results indicate that the redox behavior of Pu cannot be described by disproportionation of Pu alone. Under the experimental conditions, the redox reactions of Pu seem to be divided into two groups, Pu(IV)aq Pu(III)aq and Pu(IV)coll Pu(V)aq Pu(VI)aq. In the Pu solution containing initially only Pu(IV), the reduction of Pu(IV) to Pu(III)aq dominates rather than the oxidation to Pu(V)aq and Pu(VI)aq. The observed two groups of reactions show the dependency of pH due to the related hydrolysis and colloid formation of Pu(IV). With increasing pH, the [Pu4+] decreases either through its hydrolysis and colloid formation (increase of Pu(IV)coll) or through its reduction (increase of Pu(III)aq). The polymer species or colloids may dissolve to Pu(V)aq through the second reaction group (increase Pu(V)aq + Pu(VI)aq). Consequently, it is observed that with increase of pH, [Pu(IV)aq] decreases, [Pu(III)aq] increases, and [Pu(IV)coll]+[Pu(V)aq]+[Pu(VI)aq] increases. This study is also performed under inert gas conditions in order to investigate the influence of dissolved oxygen on the oxidation of Pu(IV) (Pu(IV)coll Pu(V)aq). From the relative abundance of the Pu oxidation states, namely the couples PuO22+/PuO2+ and PuO2+/Pu(IV)coll, the redox potential Eh(V) can be obtained. The respective values agree well with the measured Eh values. In order to use the redox couple Pu4+/Pu3+, one has to take into account the strong hydrolysis of Pu(IV) which sets in below pH 1. When the abundance of Pu4+ is calculated from the amount of by use of hydrolysis constants from earlier solvent extraction stud-ies, deviations from the measured Eh arise. By use of slightly lower values for log β1y (y = 1-2) a good agreement between all calculated and measured Eh values is achieved, suggesting that at least the first and second hydrolysis constants should be corrected.